Guest Editorial: Procedural Content Generation in Games

C REATING games is in essence a nontechnical task and, accordingly, creating computer games involves much more than just programming. First, there needs to be a game design. And then, depending on the game genre, there needs to be character models, maps, levels, boards, racing tracks, trees, rocks, weapons, textures, sound effects, quests, clouds, and so on. In many contemporary game productions, creating all this game content requires a significantly larger effort and expense than the actual programming of the game. Therefore, having access to tools that automate some of this content production would be of great and direct benefit to game developers. In addition, such techniques would also indirectly benefit game scholars, as advances in generative techniques are likely to improve our understanding of the content that is designed and of design methods in general. Fortunately, there are many methods (originating, e.g., from AI, computational intelligence, computer graphics, modeling, and discrete mathematics) that are eminently applicable to generating a large variety of game content. Conversely, efficiently generating proper game content poses a number of interesting challenges to such methods, which can spur the development of new algorithms. Therefore, the field of procedural content generation (PCG) covers a wide set of problems and methods that are of high interest to both game developers and academic researchers. The call for this special issue mentioned many different aspects of PCG for games, which was clearly reflected in a total of 22 manuscripts being submitted to the IEEE TRANSACTIONS ON COMPUTATIONAL INTELLIGENCE AND AI IN GAMES (TCIAIG) targeted at this special issue. Between November 2010 and June 2011, all these manuscripts underwent a thorough peer-review process, leading to considerable improvement and sharpening of their contributions. Eventually, the present eight papers stood out as specially representative of the current state of the field and were selected for inclusion in the special issue.